Posted in Geotechnical Engineering

Huge Sinkhole in Atacama Desert —

A sinkhole approximately 32 meters in diameter and roughly 200 meters deep opened up in the Atacama Desert of Chile, near the Alcaparrosa copper mine at the beginning of August 2022. Anytime something like this […]

Huge Sinkhole in Atacama Desert —
Posted in Geotechnical Engineering

Airfield Design and Maintenance Documents and Software Now Updated — Chet Aero Marine

It’s been a while but we’re updating our free offerings on these subjects. First, the design and maintenance documents: evidently the U.S. military learned a few things in Iraq and Afghanistan, these are fairly extensive and they can be found on these two pages: Airfield Planning and Design Airfield Inspection and Maintenance Some of the documents were […]

Airfield Design and Maintenance Documents and Software Now Updated — Chet Aero Marine
Posted in Geotechnical Engineering

Some Updates for Our Marine Offerings — Chet Aero Marine

There’s a lot going on with this site, but one thing that needs some “catching up” is our marine construction collection, which really hasn’t received a comprehensive update since this site migrated to WordPress in 2016. We’ve added a number of documents to our collection: these are as follows (in no particular order): Coatings and […]

Some Updates for Our Marine Offerings — Chet Aero Marine
Posted in Geotechnical Engineering, Soil Mechanics

The Problem of Size: Gazetas and Stokoe (1991) Revisited

In a recent exchange Dr. Mark Svinkin, who has contributed several well-read articles to this site, pointed out that he had commented on a paper by Gazetas and Stokoe (1991.) The paper, Dr. Svinkin’s comments and their response can be found here:

Although this research was done a long time ago, it’s worth revisiting because of the issue that Dr. Svinkin brings up: the issue of size, that it’s not a straightforward business to extrapolate the results of model tests in controlled environments to full-scale foundations in actual stratigraphies.

In my fluid mechanics laboratory course, I discuss the issue of dynamic similarity, how one can take an airfoil or other flying object on a small scale and, using things such as the Reynolds Number, extrapolate those results to full-scale aircraft. This has proven very useful in the development of aircraft, especially before (and even long after) the development of simulation using computational fluid dynamics.

With geotechnical engineering, it has not been quite as simple. Attempts to use things such as centrifuge testing have not been as successful as, say, wind tunnel testing has been for aeronautics. Part of the problem is, as I like to say, that geotechinical engineering is not non-linear in the same sense as fluids are. Another problem is that the earth is not as homogeneous as the atmosphere, even when altitude and weather effects are considered (and these influence each other in the course of events.) But underneath all of this there are some fundamental issues that have complicated the issue of foundation size, and Dr. Svinkin points this out. My intent is to amplify on that and remind people that these issues are still relevant.

Dr. Svinkin points out the following figure from Tsytovich (1976.) I’ve referenced this text in several recent posts. Tsytovich looks at many problems in soil mechanics differently from our usual view in this country, and his perspective is frequently insightful. (An excellent example is here.) In this diagram he shows the effect of basic foundation size on the settlement of the foundation, and Tsytovich’s own explanation of this follows:

Relationship between settlement of natural soils and dimensions of loading area (from Tsytovich (1976).) The variable F is the area of the foundation, thus the square root of F is the basic dimension of the foundation and, in the case of square foundations, the exact dimension b or B (see below.)

Thus, Fig. 90 shows a generalized curve of the average results of numerous experiments on studying the settlements of earth bases (at an average deg­ree of compaction) for the same pressure on soil but with different areas of loading. Three different regions may be distinguished on the curve: I —the region of small loading areas (approximately up to 0.25 m2) where soils at average pressures are predominantly in the shear phase, with the settlement being reduced with an increase of area (just opposite to what is predicted by the theory of elasticity for the phase of linear deformations); II — the region of areas from 0.25-0.50 m2 to 25-50 m2 (for homogeneous soils of medium density, and to higher values for weak soils), where settlements are strictly proportional to \sqrt{F} and at average pressures on soil correspond to the compaction phase, i.e., are very close to the theoretical ones; and III — the region of areas larger than 25-50 m2, where settlements are smaller than the theoretical ones, which may be explained by an increase of the soil modulus of elasticity (or a decrease of deform ability) with an increase of depth. For very loose and very dense soils these limits will naturally be somewhat different.

The data given can be used for establishing the limits of applicability of the theoretical solutions obtained for homogeneous massifs to real soils, which is of especial importance in developing rational methods of calculation of foundation settlements.

From Tsytovich (1976)

Although much of the discussion centred on Tsytovich and Barkan (1962,) there is evidence elsewhere to underscore this problem, which Tsytovich sets forth in a very succinct manner.

To begin with, let us consider the basic equation for elastic settlement, which was discussed in this post, and is as follows:

s = \frac{\omega p b (1-\nu^2)}{E} (1)


  • s = settlement of the foundation at the point of interest
  • \omega = I = influence factor, given in the table below
  • p = uniform pressure on the foundation
  • b = B = smaller dimension of rectangle or dimension of square side
  • \nu = Poisson’s Ratio of the soil
  • E = Modulus of elasticity of the soil

The values of \omega are shown below.

Similar values can be found in both the Soils and Foundations Manual and NAVFAC DM 7.

It is clear that, once one is past the basic soil properties and the pressure applied on the foundation, the settlement is proportional to the basic dimension of the foundation, which is exactly what is taking place in Region II. This is also why the bulk modulus of the soil is not a basic soil property, as I discuss in this lecture. When we consider plate load tests, we must correct them for the difference between the size of the test plate and the size of the foundation, as this slide presentation shows.

Since we are dealing with foundation dynamics, one item that seems to have fallen out of the whole discussion is that of Lysmer (1965). Lysmer’s Analogue, which reduces the response of a soil under the foundation to a simple spring-damper-mass system, defines the spring constant as follows:

K = \frac{4Gr}{1-\nu} (2)

where K is the spring constant of the soil and r is the foundation’s radius. If we break it down further, as is done in Warrington (1997,) and develop a unit area spring constant under the foundation, we have

k = \frac{4Gr}{F(1-\nu)} (3)

where k is the equivalent unit area spring constant under the soil. Equation (3) in particular shows that, for a given unit load on a foundation, the static portion of the reaction is inversely proportional to the basic size of the foundation. (The unit damping constant is actually independent of the area for round foundations.)

These results show that, while the effect of size may differ from one model to the next, it cannot be overlooked in any attempt to extrapolate physical model tests of any kind to actual use. This effect is further complicated by variations in shear modulus due to either strain softening, layered stratigraphy, effective stresses or other factors. The effect of the stratigraphy is further magnified by the fact that larger foundations have larger “bulbs of influence” into the soil and thus layers that smaller foundations would not interact with become significant with larger ones.

“Sand box” tests have other challenges. While they attempt to simulate a semi-infinite space, reflections from the walls of the box are inevitable, especially with periodic loads such as were present in this test. These challenges were documented in the original study. (An interesting study using another one of these boxes is that of Perry (1963).)

The failure of geotechnical engineering to adequately resolve the size issue, both in terms of design and in terms of using laboratory data to simulate full-scale performance, remains a frustration in geotechinical engineering. Hopefully other types of models will help move things forward, along with advances in our understanding of soil behaviour and our ability to replicate it both experimentally and numerically.

Posted in Academic Issues, Geotechnical Engineering

Explicit and Implicit Methods and Plasticity: A Warning for Code Writers and Users

It’s time to put a wrap on our series on constitutive equations, both elastic and plastic. We’ll do this in a more qualitative way by pointing out a trap in code writing that can have an effect on the results.

We generally divide problems into two kinds: static and dynamic. With static elastic problems, we can solve the equations in one shot, although we have to deal sometimes with geometric non-linearity (especially when elastic buckling is involved.) With dynamic problems time stepping is involved by definition. When we include plasticity, because of the path dependence issue we’re always using some kind of stepping in the solution. That stepping takes at least two levels: the stepping like we discussed in our last post and load and/or displacement stepping as the load or displacement is increased. Therefore, in a problem involving plasticity, we are stepping the problem one way or the other.

But how large do the steps need to be? In the case of dynamic stepping, that problem is wrapped up with the whole question of implicit vs. explicit methods. To make things simple, explicit methods are those which take information from the past and use that in each step to predict the state at the end of the step, while implicit methods use information from both start and finish of the step to predict the step end state. Implicit methods have the advantage of allowing the model to have larger time steps, which is one reason why they’ve gained currency with, say, computational fluid dynamics models.

Probably the most popular dynamic method used for analyses such as this is Newmark’s Method, which admits both static and dynamic implementations. An example of Newmark’s Method for a simple system (a ram/cushion/pile system) can be found here. In that analysis, except for inextensibility issues (the cap not sticking to the pile, for example) the analysis was entirely elastic.

It would seem that using an implicit method would be the optimal solution for a dynamic system. However, that runs into a serious problem, as discussed by Warrington (2016):

Consider the elasto-plastic model as depicted (above). At low values of strain, elasticity applies and the relationship between stress and strain is determined by the slope of the line, the modulus of elasticity. In the elasto-plastic models considered, the reality is that the relationship between stress and strain is always linear; the key difference between the elastic and plastic regions is that, upon entrance into the plastic region, there are irrecoverable strains which take place. Cook, Malkus and Plesha (1989) observe that, in the plastic region, there is a plastic modulus, which is less than the elastic modulus and, in the case of softening materials, actually negative. They also observe that, in this region, the acoustic speed is lower than that in the elastic region…This is a similar phenomenon to that of the variations in elastic modulus and acoustic speed based on strain, which complicated the determination of the applicable soil properties.

With a purely elasto-plastic model, the plastic modulus is zero, and thus the acoustic speed is also zero. This effectively decouples the mass from the elasticity in the purely plastic region. This result is more pronounced as the time (and thus the distance) step is increased; the model tends to “skip over” the elastic region and the inertial effects in that region. Thus with larger time steps inertial effects are significantly reduced, and their ability to resist pile movement is likewise reduced.

For problems such as this, the best solution is to use an explicit method with very small time steps to “catch” all of the effects of the plasticity. One major advantage of that approach is that explicit methods allow us to dispense with assembling the global stiffness matrix, giving rise to “matrix-free” methods you hear about. That significantly reduces both memory requirements and computational costs in a model. It has also led to changing even problems we consider as “static” (like static load testing) to dynamic problems with small time steps, no stiffness matrices and long run times, an acknowledgement that, strictly speaking, there are no real “static” problems, a fact that makes many civil engineers freak out.

Chances are that the effects of this have been built into whatever code you might be using, and that you’re not aware of how your code is handing it. The wise user of numerical modelling solutions would do well to familiarise him or herself of how the code being used actually handles problems such as this and becomes a more informed user of the tools at hand.